♾ Meta: AI Killed The Video Star
From content creators to an AI-driven feed
Welcome to the Free edition of How They Make Money.
Over 160,000 subscribers turn to us for business and investment insights.
In case you missed it:
“Video killed the radio star.”
In 1979, The Buggles' iconic song lamented how television and video overshadowed older media, leaving the radio star to fade into nostalgia. Now, nearly fifty years later, AI may be setting up a similar transformation, potentially killing the video star. With video currently consuming the bulk of user time on Meta’s apps, AI’s rapid evolution could redefine what people see and engage with online.
Last week, Meta reported its Q3 earnings, and the story was familiar. Rising AI investments clouded an otherwise impressive quarter. With over 3.2 billion people using at least one of Meta’s apps daily, the company holds a unique position—alongside Google and YouTube—to lead AI into the mainstream. But this evolution may come at a cost: it could fundamentally disrupt the creator economy as we know it.
So, what does it mean for the future of Meta’s apps?
Let’s visualize the quarter and get to the latest insights.
Today at a glance:
Meta Q3 FY24.
Recent business highlights.
Key quotes from the earnings call.
The death of the creator economy.
1. Meta Q3 FY24
Meta operates across two business segments:
💬 FoA: Family of Apps (Facebook, Instagram, Messenger, and WhatsApp).
🥽 RL: Reality Labs (virtual reality hardware and supporting software).
FoA Daily Active People grew +5% year-over-year to 3.29 billion. Net additions have slowed, with Meta adding around 20 million daily users in Q3 2024—down from 50 million per quarter in early 2024.
Meta’s reach now covers over half of the global population aged 15 to 80. Given this saturation, future growth will depend more on boosting engagement and ad efficiency than adding new users.
Zuck called out:
Facebook: Positive trends among Gen Z in the US.
Instagram: Sustains “strong” global growth.
WhatsApp: Surpassed 2 billion calls daily.
Meta AI: Now has 500 million monthly actives.
Threads: Reached 275 million monthly actives, up from 200 million in Q2 and growing significantly in regions like the US, Taiwan, and Japan (not yet monetized, and likely won’t contribute meaningful revenue in 2025.)
Advertising metrics:
Ad impressions grew +7% Y/Y (vs. +10% Y/Y in Q2).
Average price per ad grew +11% Y/Y (vs. +10% Y/Y in Q2).
Average revenue per person grew +12% Y/Y to $12.29 (+14% Y/Y in Q2), well above competitors like Snap ($3.10) and Reddit ($3.58).
Critics suggest Meta’s numbers might be inflated by bot activity, but ARPU growth suggests genuine ad value. You can fake users but not revenue.
Income statement:
Revenue grew +19% Y/Y to $40.6 billion ($0.3 billion beat).
FoA grew +19% Y/Y to $40.3 billion.
RL grew +29% Y/Y to $0.3 billion.
Gross margin was 82% (-1pp Y/Y, +1pp Q/Q).
Operating margin was 43% (+2pp Y/Y, +5pp Q/Q).
FoA’s operating profit was $21.8 billion (54% margin, +2pp Y/Y).
RL’s operating loss was $4.4 billion (vs. $4.5 billion in Q2).
EPS grew +37% Y/Y to $6.03 ($0.74 beat).
Cash flow:
Operating cash flow was $24.7 billion (61% margin, +1pp Y/Y).
Free cash flow was $15.5 billion (38% margin, -2pp Y/Y).
Balance sheet:
Cash, cash equivalent, and marketable securities: $71 billion.
Long-term debt: $29 billion.
Guidance:
Q4 FY24 revenue $46.5 billion in the mid-range ($0.3 billion beat).
FY24 expenses $96-$98 billion (previously $96-$99 billion).
FY24 Capex $38-40 billion (previously $37-$40 billion).
So what to make of all this?
Revenue grew 20% in constant currency, compared to 23% in Q2. The slowdown was primarily due to tougher comps (similar to YouTube).
Ad revenue growth was price-driven. Ad demand stayed strong thanks to higher ad performance, with the online commerce, healthcare, and entertainment categories contributing the most. Regarding geographic growth, North America and Europe led with 21% growth, while Asia decelerated from 28% to 15%.
Reality Lab’s revenue grew 29%, driven by hardware sales, though the segment continues to post substantial operating losses. The visual below shows Meta’s operating profit by segment in the past four years, from the COVID high to the post-ATT hangover, followed by the ‘Year of Efficiency.’ FoA operating profit is at an all-time high, and RL’s operating loss has been stable at around $4 billion quarterly.
Headcount grew 9% from a year ago and 2% sequentially to 72,404, showing a return to hiring after the year of efficiency, particularly in priority areas like monetization, infrastructure, Reality Labs, and gen AI.
Stock buybacks were nearly $9 billion in Q3, a step up from $6 billion in Q2 but down from the $15 billion in Q1. Still, it was a considerable increase compared to less than $4 billion a year ago. Management still finds the stock attractive enough. They also paid $1.3 billion in dividends (same as Q1 and Q2).
Capital expenditures rose 36% to $9.2 billion. That compares to $8.5 billion in Q2. Guidance for the full year remained in line with expectations. Management expects a “significant acceleration in infrastructure expense” in 2025. These investments will eventually flow through the P&L under the cost of revenue and R&D expenses.
Meta remains a cash-printing machine, with nearly $52 billion in free cash flow in the past 12 months, even after the heavy AI spending. For comparison, Alphabet generated $56 billion over the same period.
Q4 FY24 revenue guidance implies a deceleration. The mid-range of guidance is a 16% growth.
Now, let's look into Meta's investments and its market position.
2. Recent business highlights
👓 Meta Orion
Meta’s Orion AR glasses represent an ambitious vision to push beyond smartphones into the future of augmented reality (AR).
Prototype, not product: Orion is a high-tech AR prototype equipped with advanced features, though production costs currently make it prohibitive for consumer release.
Advanced AR display: Using Micro LED projectors and silicon carbide lenses, Orion provides a wide field of view and sharper visuals than most existing AR glasses.
Interactive AI capabilities: Orion’s integration with Meta’s generative AI enables users to interact with virtual elements, recognize real-world objects, and create on-the-spot solutions, like recipes.
Hardware complexity: The device requires a neural wristband for control and a wireless compute puck for functionality, creating a multi-part system.
High cost, limited production: With a price tag estimated at $10,000 per unit, Orion is not yet ready for mass production. Meta has produced around 1,000 units for demos and internal testing.
Future vision: Meta aims to release a consumer-ready AR device within a few years, targeting a slimmer, more affordable product closer to smartphone price levels.
Orion highlights Meta’s ambition to lead the next computing wave, though significant technological and cost hurdles remain.
Why reveal it now? Zuck is likely engaging shareholders with early prototypes to justify Reality Labs’ annual $16-20 billion operating loss and gather feedback to refine the product.
However, Meta’s early AR unveiling may come with competitive trade-offs. According to Bloomberg, Apple has launched its own smart glasses initiative, a market study called “Atlas,” signaling a potential shift from its high-end $3,500 Vision Pro VR headset. Apple recently cut its Vision Pro shipment target to less than half a million units in the first year—down from an initial target of 3 million.
🤖 How AI is already impacting Meta
Beyond long-term projects like Orion, Meta’s AI advancements are already reshaping its core business across two strategic growth levers: engagement and monetization.
📱 Engagement: Meta’s recommendation engine selects the most relevant content—primarily video—to populate user feeds, which keeps people engaged. AI’s role here is to refine prediction systems, allowing users to spend even more time on the apps by surfacing content that maximizes scrolling.
💵 Monetization: AI enhances ad efficiency across the entire lifecycle—creation, performance, and measurement. Gen AI can assist with ad copy, image, and video production, while new models can analyze user actions before serving specific ads, ultimately increasing conversions at the margins.
Another critical tool here is Meta AI Studio, which empowers developers to create, train, and deploy custom AI models across Meta’s ecosystem. By enabling partners to develop personalized assistants, interactive AI, and AR applications, Meta aims to drive the next wave of consumer apps and maximize ad potential across its platforms.
📊 Market share
Meta’s advertising revenue reached $39.9 billion in Q3, representing 81% of Google’s Search revenue, up from 76% last year. Meta has steadily closed the gap, with its ad revenue growing at the same pace as Amazon’s despite its already larger base. In short, Meta is regaining market share and continues to adapt effectively to the post-ATT landscape.
3. Key quotes from the earnings call
CEO Mark Zuckerberg
On AI and the family of apps:
“Improvements to our AI-driven feed and video recommendations have led to an 8% increase in time spent on Facebook and a 6% increase on Instagram this year alone. More than a million advertisers used our GenAI tools to create more than 15 million ads in the last month, and we estimate that businesses using Image Generation are seeing a 7% increase in conversions -- and we believe that there is a lot more upside here.”
These engagement gains are meaningful for Meta’s top line. While not all growth in impressions translates directly to revenue, AI-driven tools are already making a significant financial impact, likely contributing billions in quarterly revenue.
On Llama 4:
“We're training the Llama 4 models on a cluster that is bigger than 100k H100s or bigger than anything that I've seen reported for what others are doing.”
Zuck recently mentioned that Meta might need hundreds of billions in computing capital expenditure over the coming years to achieve its gen AI ambitions. However, as current investments in AI continue to yield engagement and revenue gains, Meta has the flexibility to scale spending as necessary based on performance.
On Rayban Meta glasses:
“I continue to think that glasses are the ideal form factor for AI because you can let your AI see what you see, hear what you hear, and talk to you. Demand for the glasses continues to be very strong. The new clear edition that we released at Connect sold out almost immediately.”
Meta is pursuing a two-pronged approach to AR glasses:
Orion has a hardware challenge (powerful but still cumbersome).
Rayban Meta glasses have a software challenge (lightweight but only offering relatively simple use cases).
CFO Susan Li:
On recommendations:
“Inspired by the scaling laws we were observing with large language models, last year we developed new ranking model architectures capable of learning more effectively from significantly larger data sets. To start, we have been deploying these new architectures to our Facebook video ranking models, which has enabled us to deliver more relevant recommendations and unlock meaningful gains in watch time.”
This advancement is key to Meta’s engagement strategy: using AI to tailor content based on each user’s unique interests. Management plans to expand these AI-driven recommendation models to all surfaces, enabling a more customized user experience and maximizing engagement across apps.
On Meta AI
“It continues to be on track to be the most used AI assistant in the world by end of year. […] A number of the frequent use cases we’re seeing include information gathering, help with how to tasks, which is the largest use case. But we also see people using it to go deeper on interests, to look for content on our services, for image generation, that’s also been another pretty popular use case so far.”
Meta’s approach to AI is clear: focus on building engagement and adoption first, with monetization to follow later. By embedding Meta AI as a versatile tool across its platforms, the company is positioning it as a go-to assistant that could potentially outpace competitors in usage.
On capital allocation:
“We are very optimistic about the set of opportunities in front of us, and believe that investing now in both infrastructure and talent will not only accelerate our progress, but increase the likelihood of maximizing returns within each area.”
Meta’s Capex has been in close range with Microsoft and Alphabet (see visual). The company’s rising operating income and free cash flow make it an easier pill to swallow for shareholders.
4. The death of the creator economy
In recent years, Facebook and Instagram have shifted from social networks (content from people you know) to content networks (content from anyone), creating unprecedented opportunities for creators. But this era may be coming to an end.
CEO Mark Zuckerberg shared:
“We’re going to add a whole new category of content, which is AI generated or AI summarized content or kind of existing content pulled together by AI in some way. And I think that that’s going to be just very exciting for Facebook and Instagram and maybe Threads or other kind of Feed experiences over time.”
In short, Meta apps are on track to become AI-driven content networks. By using AI to generate or repackage the most engaging content, Meta may no longer rely on individual creators to drive engagement.
As AI learns to replicate what’s trending and maximize engagement, creators could struggle to compete with these models, which will know exactly what to post and when. Content creators may find themselves on the sidelines, becoming virtual Uber drivers in an online world of self-driving cars. This shift could take years to fully materialize, but it’s clear there’s no turning back. With 57% of Gen Z1 aspiring to become influencers, many might face a jarring reality.
Why this matters:
YouTube pays creators a 55% share of ad revenue, but Meta doesn’t share ad revenue with creators, so AI-generated content wouldn’t be a cost optimization. Instead, it would allow the company to embed ads more seamlessly into algorithmic feeds, potentially boosting both ad impressions and conversions across its own content.
While AI-generated feeds might sound dystopian, the biggest accounts already use bot-like engagement-farming tactics. So, when AI starts to dominate our feeds, we might not even notice the difference.
That’s it for today!
Stay healthy and invest on!
Group Subscription
Premium articles like this one are intended for a single recipient. If you want to order multiple subscriptions for your team with a group discount, please contact me directly (reply to any of my emails).
Disclosure: I am long AAPL, AMZN, GOOG, and META in App Economy Portfolio. I share my ratings (BUY, SELL, or HOLD) with App Economy Portfolio members.
Author's Note (Bertrand here 👋🏼): The views and opinions expressed in this newsletter are solely my own and should not be considered financial advice or any other organization's views.
Morning Consult Survey 2023.